Fast demand response control strategy for buildings responding to urgent requests of smart grids

Abstract

The thesis presents an effective control strategy for fast demand response of buildings responding to urgent requests of smart grids. Shutting down part of operating chillers in an air-conditioning system is proposed to satisfy the urgent requirements of smart grids, i.e., immediate power reduction. Three schemes are developed for this direct load control strategy, including: power demand optimization, system sequence control resetting and online control/regulation. Moreover, an online model-based optimal control strategy is developed for achieving a best possible indoor air temperature and humidity when the proposed direct load control strategy is adopted for fast demand response. These new developed control schemes are tested and validated on a dynamic simulation platform. For the scheme of power demand optimization, the reduced building cooling demand and the power limiting threshold considering the limit of indoor air temperature increase are determined. For the scheme of system sequence control resetting, the numbers of retained operating devices in an air-conditioning system are optimized based on the reduced building cooling demand predicted by the scheme of power demand optimization. For the scheme of online control/regulation, two modules are included, i.e., the chiller load regulator and the cooling distributor. After shutting down part of operating chillers, chiller load regulator is used to adjust the total power consumption of the retained devices (i.e., chillers, pumps and fans) to follow the pre-determined power limiting threshold. The cooling distributor, which is based on a completely new control strategy developed in this thesis (i.e., supply-based feedback control strategy), is responsible for allocating the limited cooling supply properly and reasonably to individual zones/rooms during the demand response period. This can effectively avoid seriously imbalanced cooling distribution and excessively over-speeded secondary pumps and air delivery fans, which are caused by the failure of conventional control strategies after simply shutting down part of operating chillers directly. In addition, the online optimal control strategy for optimizing the air flow setting of each air handling unit (AHU) is developed under a pre-determined power limiting threshold. This control strategy benefits the proposed direct load control strategy to achieve a best possible indoor environment in terms of temperature and humidity without the need of extra power consumed. The optimal air flow settings of individual AHUs are optimized based on a model-based online optimal control strategy using genetic algorithm (GA). Furthermore, the supply-based feedback control strategy is applied for solving the problems of air-conditioning systems during morning start period. During this period, the different cooling-down speeds of indoor spaces result in the time duration for precooling the indoor spaces back to their set-points extended significantly. The extended precooling time causes serious energy waste. High peak demand during morning start period always appears because of the failure of feedback control strategies. An optimal control strategy based on the supply-based feedback control is proposed to achieve proper cooling distribution among individual zones/rooms as well as to optimize the number and schedule of operating chillers so that the precooling time duration, the energy waste and the high peak demand can be reduced effectively.